This is a very good start with respect to miniaturization... but there are several things readers need to keep in mind... a lot of things actually, for acceptance in the USA in clinical environments as an approved medical device for patient care. There are many short comings with the current device, and the likelihood of acceptance in the USA (both FDA and product market acceptance) would be extremely questionable. However, there would be a good application for this technology in the home monitoring, military and in athletics, where basic EKG data would be nice to have, just not as diagnostic (this already exists for such purposes using ANT+ in common basic physiological monitoring equipment.. watches and HR/breathing), but is very cheap and easy to implement... but WILL NEVER BE USING IN CLINICAL SETTINGS IN THE USA.
First... a little bit about ECGs (EKGs in the US) and sensors. One of the primary disadvantages of only using one electrode to detect the minute electrical signals of the heart is that those signals are highly directional. In fact, there are 12 general directions monitored in what are called 12 Lead EKGs (the gold standard... however, most real-time monitoring is of only one "lead", which is associated with one of the 12 general directions. Unfortunately, using just a single leads is not going to give any "sense of direction" and would be just a simple summary of the electrical activity all the leads. This impedes the ability of the intelligent analysis systems to determine arrhythmia changes (changes to the baseline EKG) beyond those specific to general time-related analysis and the application of chaos theory to detect sudden cardiac death for example, which is well documented in the medical literature. The primary 6 leads, are called limb leads, and are used for basic cardiac monitoring and provide longitudinal electrical "axis" data and better wave form analysis for different "parts" of an EKG. These are the classic 3 or 4 wire "patches" seen on the 4 corners of the chest. 3 wire systems (3 patches) can only record a single limb lead. All 4 patches or limb attachments (back in the old days the "limb" leads were actually your arms and legs) must be present to record all 6 limb leads... which provide about 85% of the diagnostic data needed to do a full cardiac electrical analysis. Note that automatic defibrillators use one limb lead for analysis (across 2 wires). The remaining 6 leads of a 12 lead EKG are called cardiac, or chest leads. They record data from immediately around the heart, to provide a more of a 3D image, as the heart doesn't lie flat in the chest and tends to hang to the left.
Since a single lead requires at least 2 wires to measure a polar electrical signal, this is the minimal requirement for defibrillators, and hence any electronic patient record system associated with defibrillation. For more advanced purposes... anything beyond defibrillation, more than one lead must be immediately accessible for recording purposes. Hence, any ICU/CCU/ER/Ambulance (clinical environment) will always use multiple leads, as that is the standard of care in the USA. So until this system can use multiple wireless sensors to capture multiple leads, and buffer that data for analysis and/or just raw transmission, this approach will not do much good in clinical settings in the USA. Not capturing all the raw data would also have significant issues in clinical settings.
Unfortunately, cardiac monitoring for the purposes of an electronic patient record, in clinical settings the USA, requires the raw EKG data from multiple simultaneous leads... true leads. Analysis of the data is by FDA certified software/hardware, after which data can be stored and accessed by "event or significant change from a baseline", which is also recorded on a regular basis.
The FDA hurdles for this device are significant... it would need to go through at least two rounds of FDA acceptance for two separate functions.... the physical device itself, and the software it uses for both intelligent data analysis and data transmission. These are significant hurdles. Physically, the device would need to tolerate repeated defibrillations, including direct discharge contact, and would need to function with external pacemakers. The 2.4 GHz range is very susceptible to noise, especially high powered WiFi... which begs the question of why use ANT+ in clinical settings, especially when you need a gateway to TCP/IP anyway.
Given what FDA approval would cost (millions), and given that the EKG wires are not really a hassle to clinical providers as long as they are not attached to a physical monitor (read... this means the wires are attached to a wireless hub that is attached to the patient and sends all lead data [not summaries or events] via TCP/IP [with full buffering] over WiFi, with IEEE 802.1 physical security certs and WPA2 Enterprise encryption) you will never see this device in a clinical setting in the USA as it doesn't meet clinical standards.
WiFi, when properly secured (e.g. with WPA2) is quite secure The reason it's inappropriate here has to do with power requirement for WiFi.
ANT is a short range protocol that is more suited to low-power app like this.
Full disclosure: I work in Nordic Semiconductor's marketing department. First, it's important to note that Lifetouch monitors are not currently being offered for commercial or clinical use in the U.S. The Isansys HRV011 is currently undergoing medical trials in a clinical environment in Europe. As in the U.S., patient confidentiality is a big deal there and transmitted data is fully encrypted. Isansys uses the ANT base protocol (see www.thisisant.com) but adds its own proprietary application layer for added security. As "Noswad" comments the only thing that's attached to the patient is the 'patch' shown in the picture at the head of the article.
I wish no one had to worry about hackers in any application, but we do. Unfortunately, data security is never a non-issue, and wireless networks are eminently hackable. At least in the US, confidentiality of patient records is a legal issue. So I'm still curious about the ANT standard and encryption.
What possible commercial or criminal interest could there be in hacking this data? If the patient has already been through intensive care, his/her insurance credentials are shot, so the insurance companies don't stand to gain from intercepting data. Data security is therefore a non-issue - simple packet ID would be enough.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.